Skip to content

[Fix] Add operations in _dummy_run to maintain synchronization with _process_reqs, resolving a service hang #2454

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Aug 25, 2025
Merged

[Fix] Add operations in _dummy_run to maintain synchronization with _process_reqs, resolving a service hang #2454

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9c5d40e
This change separates the general DP padding logic from the existing …
yiz-liu Aug 20, 2025
e659504
Merge branch 'vllm-project:main' into fix-dp
yiz-liu Aug 23, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions vllm_ascend/worker/model_runner_v1.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1914,6 +1914,10 @@ def _dummy_run(
)

# Padding for DP
num_pad, num_tokens_across_dp_native = self.get_dp_padding(num_tokens)
# num_tokens += num_pad ## Uncomment this after TorchAir is removed

# Padding for DP (for TorchAir)
Copy link
Collaborator

@weiguihua2 weiguihua2 Aug 23, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this _get_forward_metadata_across_dp_and_pad function not for torchair, this method can be rewritten in torchair runner

Copy link
Collaborator Author

@yiz-liu yiz-liu Aug 25, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sure, will figure out a way to merge these two paths.

(num_tokens, num_tokens_across_dp, with_prefill,
_) = self._get_forward_metadata_across_dp_and_pad(
num_tokens, with_prefill, False)
Expand Down
Loading